I have created a web-project and deployed it to a tomcat web-server on a windows machine. Now I have a hadoop cluster running on a separate Centos machine. I have followed the standard word-count example to develop my code and run through eclipse. I have followed descriptions given in below links.
To run the program : http://stackoverflow.com/questions/9849776/calling-a-mapreduce-job-from-a-simple-java-program
To change username to Centos machine user : http://stackoverflow.com/questions/11041253/set-hadoop-system-user-for-client-embedded-in-java-webapp
But this does not work for some reasons. If I package my code in a jar and run on hadoop machine, it runs fine, but if I run on eclipse windows machine, nothing happens. If someone could point me to a tutorial for this, that would be great. Thanks.
